Robust penalized extreme learning machine regression with applications in wind speed forecasting

Yang, Yang; Zhou, Hu; Gao, Yuchao; Wu, Jinran; Wang, You‐Gan; Fu, Lili

doi:10.1007/s00521-021-06370-3

Cited by 20 publications

(9 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compare the proposed method with ELM [1], WELM [11], IRWELM [12], Welsch-ELM [22], Laplace-ELM [25], p-Welsch-ELM [28] on the artificial datasets and benchmark datasets. The root mean square error (RMSE) is chosen as the evaluation metric:…”

Section: Methodsmentioning

confidence: 99%

“…However, many practical applications cannot guarantee the error followed a normal distribution, which lead to a fact that ELM is highly susceptible to noise and outliers. Subsequently, the researchers proposed several loss function such as Huber [11], l 1 [12] and Pinball [13] and their corresponding ELM models. However, these loss functions were still less robust because they had a linear relationship with the training error and increased linearly with the training error.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust Extreme Learning Machine Based on p-order Laplace Kernel-Induced Loss Function

Luo,

Wang,

Lin

2024

IJACSA

View full text Add to dashboard Cite

Since the datasets of the practical problems are usually affected by various noises and outliers, the traditional extreme learning machine (ELM) shows low prediction accuracy and significant fluctuation of prediction results when learning such datasets. In order to overcome this shortcoming, the l2 loss function is replaced by the correntropy loss function induced by the p-order Laplace kernel in the traditional ELM. Correntropy is a local similarity measure, which can reduce the impact of outliers in learning. In addition, introducing the p-order into the correntropy loss function is rewarding to bring down the sensitivity of the model to noises and outliers, and selecting the appropriate p can enhance the robustness of the model. An iterative reweighted algorithm is selected to obtain the optimal hidden layer output weight. The outliers are given smaller weights in each iteration, significantly enhancing the robustness of the model. To verify the regression prediction of the proposed model, it is compared with other methods on artificial datasets and eighteen benchmark datasets. Experimental results demonstrate that the proposed method outperforms other methods in the majority of cases.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Robust Extreme Learning Machine Based on p-order Laplace Kernel-Induced Loss Function

Luo,

Wang,

Lin

2024

IJACSA

View full text Add to dashboard Cite

show abstract

“…Therefore, to make more targeted recommendations and suggestions with guidance for the DMUs, an effective influence factor analysis method is needed for analysis. Common analysis methods include the least-square method [40], stochastic frontier method [41], Lasso method [42,43], robust penalized extreme learning machine regression [44], and physics-informed statistical learning method [45]. However, the efficiency value measured by the DEA method has a lower limit, and using traditional analysis methods is likely to cause serious bias.…”

Section: Influencing Factors Analysis Methodsmentioning

confidence: 99%

Spatial-Temporal Pattern and Influencing Factors of Ecological Efficiency in Zhejiang—Based on Super-SBM Method

Huang

Zhang

2022

Environ Model Assess

View full text Add to dashboard Cite

The traditional meaning of ecological efficiency generally considers only the ratio of economic output to environmental input. This paper expands the meaning and the evaluation system of ecological efficiency from the perspective of improving people's livelihoods. Not only are the discharge of wastewater, waste gas, and solid waste included in the undesired output, but the output index also takes full account of the overall development of the economy, innovation, society and the environment from the perspective of high-quality development. Under the assumption of variable returns to scale, a super-efficiency slack-based measure model based on the undesirable output and Malmquist index is introduced to measure the spatial and temporal variation of ecological efficiency of Zhejiang Province in China, and the panel Tobit method is used to study the key factors affecting ecological efficiency. The results include the four following findings: (1) In the past 12 years, the ecological efficiency of Zhejiang Province has steadily increased, except in 2019 and 2020, when seven cities in Zhejiang Province experienced a decline or near stagnation due to the impact of the economic slowdown and the COVID-19 epidemic. (2) The ecological efficiency of Zhejiang demonstrates a severe regional imbalance, showing a high level in the northeast and a low level in the southwest. (3) Malmquist index analysis shows that the improvement of ecological efficiency in Zhejiang Province has shifted from mainly relying on the dual drivers of pure technical efficiency and scale efficiency in the early stage to relying on technological progress in the later stage. (4) Tobit regression analysis shows that industrialization structure, Theil index, and traffic activity have a significant positive effect on ecological efficiency.

show abstract

“…Yang et al (2022a) developed a highly accurate short-term load forecasting method using non-linear auto-regressive artificial neural networks with exogenous multi-variable input. Yang et al (2022b) presented a novel approach for shortterm electrical load forecasting by the radial basis function neural networks, and the result showed that the application of neural networks in short-term load forecasting is encouraging. Wang et al (2016) proposed an outstanding model based on a wavelet neural network to address the complex nonlinearities and uncertainties in forecasting the electric load, and the accuracy of the proposed model is better than the considered models.…”

Section: Introductionmentioning

confidence: 99%

A hybrid SVR with the firefly algorithm enhanced by a logarithmic spiral for electric load forecasting

Zhang

Shi

et al. 2022

Front. Energy Res.

Self Cite

View full text Add to dashboard Cite

Accurate forecasting of an electric load is vital in the effective management of a power system, especially in flourishing regions. A new hybrid model called logarithmic spiral firefly algorithm-support vector regression (LS-FA-SVR) is proposed to promote the performance of electric load forecasting. The new hybrid model is acquired by combining the support vector regression, firefly algorithm, and logarithmic spiral. Half-hourly electric load from five main regions (NSW, QLD, SA, TAS, and VIC) of Australia are used to train and test the proposed model. By comparing the model results with observed data on the basis of the root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percent error (MAPE), the performance of the proposed hybrid model is the most outstanding among all the considered benchmark models. Hence, the results of this study show that the hybrid model LS-FA-SVR is preferable and can be applied successfully because of its high accuracy.

show abstract

Robust penalized extreme learning machine regression with applications in wind speed forecasting

Cited by 20 publications

References 59 publications

Robust Extreme Learning Machine Based on p-order Laplace Kernel-Induced Loss Function

Robust Extreme Learning Machine Based on p-order Laplace Kernel-Induced Loss Function

Spatial-Temporal Pattern and Influencing Factors of Ecological Efficiency in Zhejiang—Based on Super-SBM Method

A hybrid SVR with the firefly algorithm enhanced by a logarithmic spiral for electric load forecasting

Contact Info

Product

Resources

About